Dentomaxillofacial Radiology (2001) 30, 141 ± 146

ã 2001 Nature Publishing Group. All rights reserved 0250 ± 832X/01 $15.00
www.nature.com/dmfr

Comparison of computed tomography with conventional

radiography for midfacial fractures
R Tanrikulu*,1 and B Erol1
1
Department of Oral and Maxillo-facial Surgery, Faculty of Dentistry, University of Dicle, Diyarbaki r, Turkey
.

Objective: To compare the clinical e€ectiveness of computed tomography (CT) with

conventional radiography in midfacial fractures.
Methods: The conventional radiographs (CM) and CT scans of 40 consecutive patients with
complex midfacial fractures were assessed independently by two examiners. The number and
site of fractures of the orbit, zygoma and maxilla were compared by the Wilcoxon Matched
Pairs Signed Rank test. The best method for classi®cation of the fracture was determined.
Results: Coronal CT (CCT) proved superior in the diagnosis of orbital fractures (P50.001).
There was no signi®cant di€erence between any of the imaging methods for fractures of the
zygoma. Axial CT (ACT) was the most e€ective method in imaging of maxillary fractures
(ACT-CM; P50.001, ACT-CCT; P50.01). CCT was the most useful in classi®cation of orbital
and maxillary fractrue.
Conclusion: CCT is superior to CM for the assessment of complex midface fractures

Keywords: tomography, X-ray computed; radiography; fractures; facial bones

Introduction

One of the important factors determining the success of
treatment of midfacial fractures is early and correct
diagnosis.1 ± 3 The evaluation of the trauma of the facial
skeleton is based on clinical examination followed by
the appropriate radiographs.4 Plane radiography has
been used successfully for many years but advanced
imaging methods such as CT and 3-D CT have been
applied more recently.5 ± 9 While these studies have
established the clinical utility of CT, comparison with
plain radiography has not been subjected to rigorous
statistical analysis to determine its clinical e€ectiveness.
The present study was therefore undertaken to remedy
this omission.
Materials and methods
This prospective study was based on 40 consecutive
patients who underwent conventional radiography
(CM) and CT for complex midfacial fractures between
*Correspondence to: R Tanrikulu, University of Dicle, Faculty of Dentistry,
. scans was read twice. The number and site of fractures
21280 Diyarbaki r, Turkey
Received 22 July 1999; accepted 11 December 2000
1995 ± 1997 at the Department of Oral and Maxillofa-
cial Surgery,
. Faculty of Dentistry, University of Dicle,
Diyarbakir, Turkey.
All the patients involved in this study were
scheduled for operation because of the severity of
their injuries. The decision to undertake CT was based
on the existence or suspicion of at least one or more
clinical ®ndings or types of fracture summarised in
Table 1.
The radiological examination consisted of lateral
facial bones, postero-anterior (PA) skull, Waters' and
submentovertex radiographs and axial and coronal CT.
However coronal CT (CCT) was not possible in nine of
the patients due to their medical condition. Conven-
tional radiographs were obtained at 75 kV and
750 mAs with a DL-121 MB-1 skull unit (Toshiba,
Tokyo, Japan) and CT scans as 5 mm sections at
250 kV and 300 mAs as bone windows with a TCT
600S Scanner (Toshiba, Tokyo, Japan). The radio-
graphs and CT scans were reported independently by a
consultant radiologist specialising in maxillofacial
imaging in Department of Radiology, Dicle University
and the ®rst author (RT). Each set of radiographs and
of the zygoma, maxilla and orbit were recorded.
 Midface fractures
R Tanrikulu and B Erol
142
Fractures were classi®ed as blow-out, tripod or Le Fort
I, II or III. Wilcoxon Matched Pairs Signed Rank test
was used to compare two methods. P50.001 was
considered very signi®cant and P50.01 signi®cant.
In addition, the best method for classi®cation of
these fractures was determined.
Results
There were no disagreements between the ®rst and
second readings of each observer or between the two
observers.
The results of the classi®cation of the fractures
present with conventional radiography, axial and
coronal CT are shown in Table 2, the total number
of fractures of the orbit, zygoma and maxilla in Table
3 and the statistical analysis in Table 4
CCT was more e€ective for determining the types and
sites of fractures, in other words in their classi®cation
(Table 2, Figure 1a,b).
Zygomatic fractures
There was no signi®cant di€erence between any of the
methods for imaging fractures of the zygoma (Table 4).
ACT proved inadequate for imaging fractures of
frontozygomatic suture but there was no di€erence
between any of the methods for fractures of
zygomaticofrontal process or zygomatic bone (Table
6).
ACT was the best method for fracture of the
zygomatic arch (P50.001). Widening of the zygoma-
ticotemporal suture was overlooked on submentovertex
views due to superimposition but were clearly seen on
ACT. Fractures of the zygoma could be classi®ed with
all modalities (Table 2, Figure 2a ± c).

Orbital fractures
It can be seen from Table 4 that CCT was the most
e€ective method for imaging orbital fractures
(P50.001). There was a signi®cant di€erence in
imaging fractures of the lateral and medial orbital
walls between both forms of CT and conventional
methods (P50.01 and P50.001 respectively) but no
di€erence between ACT and CCT (Table 5). CCT was
signi®cantly better than both ACT and CM for the
infra-orbital rim and orbital ¯oor fractures (P50.01
and P50.001 respectively). In addition, we found that
Maxillary fractures
ACT was the most e€ective method for imaging of the
maxilla (P50.001) (Table 4). It was the most useful
method for imaging of anterior and posterior walls of
the maxillary sinus (P50.01) (Table 7). Both ACT and
CCT were signi®cantly better for the lateral and medial
walls (P50.01) (Table 7). ACT could also demonstrate
fractures of pterygoid plate.
Fractures of the maxilla were better classi®ed by
CCT (Table 2). An example of a Le Fort I fracture

Table 1 The indication for CT scanning used in this study on the

basis of the clinical and conventional radiological findings Table 3 Number of fractures by anatomical site in 40 patients
1 Suspected blow-out fractures Site CM ACT CCT
2 Fractures of zygomaticomaxillary complex
Orbit 53 86 139
3 Le Fort II or Le Fort III fractures
Zygoma 83 77 88
4 Serious facial trauma with multiple facial structures
Maxilla 73 251 192
5 Diplopia and/or enopthalmos
6 When good quality plane radiographs cannot be obtained due to CM: plain radiography; ACT: axial CT; CCT: coronal CT
the medical condition of patient

Table 4 Comparison of the effectiveness of the three imaging

modalities for determining the anatomical location of midfacial
Table 2 Classification of facial fractures with each imaging modality fractures
Classification CM ACT CCT Fracture
Blow-out 2 3 9 Orbit Zygoma Maxilla
CM ACT CM ACT CM ACT
Tripod 23 23 24
Le Fort I 1 4 7 ACT 0.000** ± 0.094 ± 0.000** ±
Le Fort II 1 0 3 CCT 0.000** 0.000** 0.250 0.206 0.000** 0.002*
Le Fort III 1 0 1 CM: plain radiography; ACT: axial CT; CCT: coronal CT. *P50.01,
CM: plain radiography; ACT: axial CT; CCT: coronal CT **P50.001

Table 5 Comparison of the effectiveness of the three imaging modalities for the diagnosis of orbital fractures in 40 patients
Orbital fractures
LOW MOW IOR OF
CM ACT CM ACT CM ACT CM ACT
ACT 0.000** ± 0.001** ± 0.020 ± 0.102 ±
CCT 0.003* 0.414 0.001* 0.248 0.009* 0.000** 0.001** 0.001**
CM: plain radiography; ACT: axial CT; CCT: coronal CT; LOW: lateral orbital wall; MOW: medial orbital wall; IOR: infraorbital rim; OF:
orbital ¯oor. *P50.01, *P50.001

Dentomaxillofacial Radiology
 Midface fractures
R Tanrikulu and B Erol
143
a which was classi®ed by the three methods is shown in
Figure 3a ± c.

Discussion

Previous workers have emphasised that the most

b been suggested that it is generally sucient to evaluate
Figure 1 (a) Coronal CT is superior to axial CT for the assessment
of a blow-out fracture since it displays the downward herniation of
orbital contents (arrow). (b) Axial CT section of the same patient
demonstrating a comminuted fracture of the orbital ¯oor
important advantage of CT in the evaluation of facial
trauma is the simultaneous evaluation of both the
brain and facial skeletion.10,11 Facial fractures, espe-
cially mid-face fractures which occur as a result of
blunt trauma, such as trac accidents or violence, are
generally accompanied by intracranial pathology or
fractures of the cervical vertebra. For this reason, the
problems in positioning such patients for CT have been
emphasised and ACT is recommended as the safest
method in such circumstances.12,13 Our ®nding that
CCT could not be obtained in nine patients is
consistent with this view.
Orbital fractures are frequently treated with bone
grafting or reconstruction plating. Radiological exam-
ination is important because of the necessity of
placing the graft or plates over sound bone.14 It has
the ¯oor and rim of the orbit from Waters' and
Caldwell projections and that the Waters' view
provides satisfactory images for upper, lower and
lateral walls espeically.15,16 However, the same
researchers have emphasised that CT should be
performed in the presence of complex orbital
injuries, since infraorbital rim fractures can be
confused with orbital ¯oor fractures in particular.15,16
In addition, many researchers have stated that the
great advantage of CT in comparison with CM is the
ability to image soft tissue so that it is possible to
evaluate the extra-ocular muscles, optic nerve and
globe12,16,17,18.
De®nitive radiological diagnosis of fractures of
medial and lateral orbital wall is important since they
may cause damage to the nasolacrimal duct or the
inferior or superior orbital ®ssure syndrome.19 Early
decompression is essential in the latter and depends on
early radiological diagnosis.19 Although Russell et al.13
proposed that the CCT is the best method for
diagnosis of medial orbital wall fractures, others have
reported that both ACT and CCT are adequate.12,19 We

Table 6 Comparison of the effectiveness of the three imaging modalities for the diagnosis of fractures of the zygoma in 40 patients
Zygomatic fractures
F-ZS Z-FP ZA Z Z-MS
CM ACT CM ACT CM ACT CM ACT CM ACT
ACT 0.000* ± 0.05 ± 0.0008* ± 1.000 ± 0.001** ±
CCT 0.32 0.000* 0.05 0.020 0.004* 0.000** 0.7 0.56 0.002* 0.32
CM: plain radiography; ACT: axial CT; CCT: coronal CT; F-ZS: frontozygomatic suture; Z-FP: zygomaticofrontal process; ZA: zygomatic
arch; Z: zygoma; Z-MS: zygomaticomaxillary suture. *P50.01; **P50.001

Dentomaxillofacial Radiology
 Midface fractures
R Tanrikulu and B Erol
144
a b

Figure 2 (a) Waters' view showing a tripod fracture of the left zygoma (arrows). (b) Axial CT of the same patient showing medial rotation of
the body of the zygoma (arrows). (c) Coronal CT con®rms the rotation of the body of the zygoma (arrows)

Table 7 Comparison of the effectiveness of three imaging modalities for the diagnosis of maxillary fractures in 40 patients
Fractures of maxilla
PB AW PW LW MW
CM ACT CM ACT CM ACT CM ACT CM ACT
ACT 0.023 ± 0.000* ± 0.001* ± 0.000* ± 0.000* ±
CCT 0.317 0.564 0.010* 0.000** 1.000 0.001* 0.000* 0.010* 0.000* 0.108
CM: plain radiography; ACT: axial CT; CCT: coronal CT; PB: palatal bone; AW: anterior wall of the maxillary antrum; PW: posterior wall of
the maxillary antrum; LW: lateral wall of the maxillary antrum; MW: medial wall of the maxillary antrum. *P50.01, **P50.001

Dentomaxillofacial Radiology

a b
Figure 3 (a) Waters' view showing a Le Fort I fracture (arrows). (b) Axial CT of the same patient showing that although the fracture lines run
parallel to the plane of section, signi®cant displacement can be seen (arrows). (c) Coronal CT of the same patient con®rming the displacment of
the maxilla (arrows)
found no di€erence between ACT and CCT in imaging
fractures of medial and lateral walls.
The importance of CCT in the diagnosis of blow-out
fractures of the orbital ¯oor has been stressed.12,14
Kassel et al.12 noted that enophthalmos may be
masked by edema or hematoma and does not present
as an early ®nding unless there is an extensive blow-
out. CT may be useful to assess the factors, such as
muscle edema or intra-orbital hematoma, which are
helpful in predicting which cases are more likely to
develop enophthalmos. CT can di€erentiate pseudo-
prolapse of orbital tissue.12 We have con®rmed the
superiority of CT (especially CCT) in the diagnosis of
fractures of the orbital ¯oor. While the Waters' view
may be sucient for fractures of infraorbital rim, CCT
Midface fractures
R Tanrikulu and B Erol
145
is indicated suspected blow-out fractures, especially in
the patients who may develop diplopia or enophthal-
mos.
There are di€erent opinions in the literature on the
evaluation of fractures of the zygoma with conven-
tional methods. Patria and Blaser16 have reported that
Waters' projection alone is not sucient to determine
the depression and rotation of zygoma and should be
supplemented with other conventional methods.
Accordingly, Da€ner et al.20 have proposed that the
lateral projection is useful in diagnosing dislocation,
rotation and depression of the molar prominence.
Johnson17 pointed out that any dislocation can be
evaluated adequately from a Waters' view by compar-
ison with the sound side.
Dentomaxillofacial Radiology
 Midface fractures
R Tanrikulu and B Erol
146
There is a consensus that conventional methods are
adequate for the evaluation of fractures of the zygoma
and CT is not required.13,17 It is only needed when there
is a gross dislocation associated with an orbital
fracture. Dislocation and rotation can be easily
displayed by both ACT and CCT.13,17 It is claimed
that CT is superior to conventional methods for the
diagnosis of zygoma fractures for two reasons; ®rst,
since the exact diagnosis of displacement of each of the
®ve major articulations of the zygoma can be evaluated
better with CT, it facilitates the selection of the best
surgical approach;14 second, depression of zygomatic
arch may trap the coronoid process of the mandible
and this complication is more easily appreciated on
ACT.19 Our ®ndings support these views. Diagnosis of
undisplaced fractures of the zygomaticotemporal suture
is important especially in the presence Le Fort
fractures when circumzygomatic wiring method is
contraindicated.
Rowe et al.11 and Russell et al.13 found that CM is
adequate for simple Le Fort fractures but in more
complex injuries, CT is required. Johnson17 stressed
References
1. Marciani DR, Gonty AA. Principles of management of complex
craniofacial trauma. J Oral Maxillofac Surg 1993; 51: 532 ± 542.
2. Mathog RH, Zachary R. Complication in the treatment of facial
fractures. Otolaryngol Clin North Am 1976; 9, 533 ± 553.
3. Stoker DJ. Skeletal trauma-acute injuries. In: Murray RO,
Jacobson HG (eds). The Radiology of Skeletal Disorders.
Volume I. London: Churchill Livingstone, 1977, pp 200 ± 207.
4. Manson PN. Management of midface fractures. In: Georgiade
GS, Georgiade NS, Riefkol R, Barwick WJ (eds). Textbook of
Plastic, Maxillofacial and Reconstructive Surgery. 2nd edn.
Volume I, Baltimore: Williams and Wilkins, 1994, pp 409 ± 432.
5. Koltai PT, Wood GW. Three dimensional CT reconstruction for
the evaluation and surgical planning of facial fractures.
Otolaryngol. Head Neck Surg 1986; 95: 10 ± 15.
6. Laine FJ, Conway WF, Laskin DM. Radiology of maxillofacial
trauma. Curr Probl Radiol 1993; 22: 145 ± 188.
7. Fuji N, Yamashiro M. Computed tomography for the diagnosis
of facial fractures. J Oral Maxillofac Surg 1981; 140: 533 ± 541.
8. Unger JM. Orbital apex fractures. The contribution of
computed tomography. Radiology 1984; 150: 713 ± 717.
9. Zilkha A. Computed tomography in facial trauma. Radiology
1982; 144: 545 ± 548.
10. Rhem GG, Ross SE. Diagnosis of unsuspected facial fractures
on routine head computerized tomographic scans in the
unconscious multiply injured patient. J Oral Maxillofac Surg
1995; 53: 522 ± 524.
11. Rowe LD, Miller E, Brandt-Zawadzki M. Computed tomo-
graphy in maxillofacial trauma. Laryngoscope 1991; 91: 745 ±
757.
12. Kassel EE, Noyek AM, Cooper PW. CT in facial trauma. J
Otolaryngol 1983; 12: 2 ± 15.
Dentomaxillofacial Radiology
that ACT and CCT should be combined, especially for
Le Fort II and Le Fort III fractures. While Kassel et
al.12 stressed particularly the e€ectiveness of CCT,
Johnson17 proposed that ACT should be preferred
since it shows the posterior dislocation in Le Fort
fractures. We have established statistically that ACT is
more e€ective than CCT for maxillary fractures, the
main reason being that the latter is ine€ective in
locating fractures of the anterior and posterior walls.
On the other hand, according to our ®ndings, CCT is
more useful than ACT for classifying maxillary
fractures.
It has been pointed out that one of the advantages of
CT in the evaluation of maxillary fractures is in
identifying fractures of the cribriform plate20 ± 23 Noyek
et al.23 have stressed the importance of early diagnosis
with CT since CSF leakage can only be treated with
reduction and ®xation of the Le Fort fracture. Some
studies have also reported that CT is superior to CM
for imaging the frontal process of maxilla, pterygoid
plate fractures and comminuted fractures of the
maxillary sinus walls.16,19,22
13. Russell JL, Davidson MJC, Daly BD, Corrigan AM. Computed
tomography in the diagnosis of maxillofacial trauma. Br J Oral
Maxillofac Surg 1990; 28: 287 ± 291.
14. Assael LA. Clinical aspects of imaging in maxillofacial trauma.
Radiol Clin North Am 1993; 31: 209 ± 220.
15. Hammerschlag SB,Hughes S, O'Reilly GV, Weber AL. Another
look at blow-out fractures of the orbit. Am J Roentgenol 1982:
139: 133 ± 137.
16. Pathria MN, Blaser SI. Diagnostic imaging of craniofacial
fractures. Radiol Clin North Am 1989; 27: 839 ± 853.
17. Johnson DH. CT of maxillofacial trauma. Radiol Clin North Am
1984; 22: 131 ± 143.
18. Gentry LR. Facial trauma and associated brain damage. Radiol
Clin North Am 1989; 27: 435 ± 446.
19. Gentry LR, Manor WF, Turski PA, Strother CM. High
resolution CT analysis of facial struts in trauma: 2: Osseous
and soft tissue complications. Am J Roentgenol, 1983; 140: 533 ±
541.
20. Da€ner RH, Apple JS, Gehweiler JA. Lateral view of facial
features. New observations. Am J Roentgenol 1983; 141: 587 ±
591.
21. Brandt-Zawadtzki MN, Minagi H, Federle MP, Rowe LD: High
resolution CT with image reformation in maxillofacial pathol-
ogy. Am J Roenterol 1982; 138: 477 ± 483.
22. Johnson DH, Colman M, Larsson S, Garner OP, Hanefee W.
Computed tomography in medial maxillo-orbital fractures. J
Comput Assist Tomog 1984; 8: 416 ± 419.
23. Noyek AM, Kassel EE, Wortzman G. Sophisticated CT in
complex maxillofacial trauma. Laryngoscope 1982; 92: 1 ± 17.